Rev. sci. tech. Off. int. Epiz., 1982, 1 (2), 387-393. Serological tests for foot-and-mouth disease in bovine serum samples. Problems of interpretation R.S. HEDGER*, I.T.R. BARNETT*, D.V. GRADWELL** and Paula TRAVASSOS DIAS*** Summary : When fully susceptible animals are infected with foot-andmouth disease for the first time the results of serological tests are usually unequivocal. Problems arise when animals have been previously infected or vaccinated with other foot-and-mouth disease virus types. The finding, in free-living African buffalo in Mozambique and South Africa, of serum neutralising titres to types O, A and Asia 1 viruses equal to or even greater than those against the SA T virus types was unexpected and led to further investigations. Although the reason for such high titres must remain speculative, it was concluded they were not the result of infection with those particu­ lar types, but were heterotypic responses following multiple infections with other prevalent types, i.e. the SAT viruses. The interpretation of serological results in the absence of a known history of disease or vacci­ nation is discussed. INTRODUCTION When fully susceptible animals are infected for the first time, the results of serum neutralisation (SN) tests against the various immunological types of foot-and-mouth disease virus (FMDV) are usually unequivocal and high titres are recorded only to the type causing the infection. Problems arise, however, when infection occurs in animals previously infected or vaccinated with other FMDV types. Attention has been drawn to heterotypic immune responses following vaccination of previously infected cattle and goats in Africa (Hed­ ger, 1966), and in an experimental study of multiple infection of cattle with FMDV, Cottral and Gailiunas (1971) recorded heterotypic neutralising anti­ body responses following successive infections with viruses of different types. They reported the heterotypic titres were generally markedly lower than the * Animal Virus Research Institute, Pirbright, Woking, Surrey, G U 2 4 ONF (U.K.). ** Coopers (South Africa) (Pty) Ltd., Rigger Road 68, Spartan, Kempton Park, 1620 (South Africa). *** Instituto Nacional de Investigaçâo Veterinaria, C . P . 1922, Maputo (Mozambique). — 388 homotypic titres but noted that cross neutralisation reactions were usually correlated with subsequent partial or complete resistance to infection. In a field study of cattle, sheep and goats of unknown disease history in the Sultanate of Oman (Hedger et al., 1980), it was found that animals undergoing natural infection with one type of FMDV (type O) had equal or even higher serum antibody titres against one or more of the other types tes­ ted (types A, C, Asia 1 and SAT 1). In a recent survey of sera from free-living buffalo (Syncerus caffer) in southern Africa, results were similarly unexpected and attempts were made to assess their significance. MATERIALS A N D METHODS In Mozambique and South Africa, buffalo were sampled during game culling operations. In Mozambique animals were shot and bled after death; in South Africa they were immobilised using projectile syringes fired from helicopters and sampled where possible before death. In Botswana buffalo were immobilised during wildlife studies and released after sampling. Blee­ ding was from the jugular vein and oesophageal/pharyngeal (O/P) samples for virus isolation were collected and handled as previously described (Hed­ ger, 1968). Virus Isolation. Virus isolation and titration was carried out on monolayers of primary bovine thyroid cells in roller tubes and the specificity of all virus strains was verified by microtitre complement fixation tests (Casey, 1965). Serum Neutralisation. Sera were assayed in virus neutralisation tests on monolayers of IB-RS2 cells in microtitre plates as described by Golding et al. (1976). The types SAT 1, 2, 3 and types O and A virus strains used in test were isolated from outbreaks of disease in cattle in southern Africa and types Asia 1 and C viru­ ses were the standard reference strains, Pak 1/54 isolated in Pakistan in 1954 and C Noville isolated in Switzerland in 1966. Serum neutralisation titres are expressed as the reciprocal of the serum dilution present in the serum/virus mixture at the 50% end point estimated according to the method of Kärber (1931). RESULTS In 1978, sera were taken from a number of free-living African buffalo shot in the Marromeu game reserve south of Maputo in the Sofala province — 389 — of Mozambique. This reserve had a large and growing buffalo population estimated at approximately 40,000 animals and culling had become neces­ sary. The sera were assayed against the Southern African Territories (SAT) FMDV types 1, 2 and 3, but, as in recent years, outbreaks of disease in cattle had occurred in Mozambique due to type O virus and, in neighbouring Malawi, to type A virus, these two virus strains, previously exotic to southern Africa, were also included in the tests. SN titres up to 1 in 355 were recorded in buffalo sera, not only to the three SAT viruses but also to types O and A in many animals (Table I). This finding was unexpected and the examination was extended to include the remaining two of the seven FMD immunological types, i.e. types C and Asia 1 viruses. Similar high titres (up to 1 in 512) were also found to type Asia 1 virus, a virus type which has never been recorded in Africa. No O/P samples for virus isolation were taken from these animals. TABLE I Reciprocal SN antibody titres in free-living African buffalo in Mozambique Animal No. Estimated age (in yrs.) SAT 1 SAT 2 Bo 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 8 6 10 2 1/2 1 3 1/2 1 1/2 5 1/2 2 1 3 4 1/2 4 1/2 5 7 6 1 1 1/2 5 1/2 8 178 256 128 355 11 512 708 355 1400 64 355 1024 355 512 708 128 355 708 32 1400 256 128 90 45 32 64 22 178 32 90 128 90 64 32 22 45 22 178 256 90 22 45 Antibody titres SAT 3 O A 178 178 256 355 178 256 1400 256 1400 90 708 1400 355 128 708 128 256 355 355 1024 708 45 90 11 16 8 90 45 64 128 11 22 178 16 32 16 16 32 22 32 45 90 22 45 22 256 22 22 355 22 178 64 90 90 8 22 32 90 128 90 22 128 22 C Asia 1 22 45 8 32 6 16 22 11 45 6 32 45 22 22 22 6 11 6 6 6 90 128 90 64 90 64 128 512 90 128 64 64 128 45 45 45 64 45 22 45 45 64 However, at the same time as this study was taking place, a similar study was being undertaken in the Kruger National Park (KNP) of South Africa which borders Mozambique. In this study O/P samples were taken and FMDV was isolated from 58 of the 73 young, 1 to 3 year old, buffalo sam- — 390 — pled in various parts of the park. Only SAT virus types were isolated (23 X SAT 1, 26 X SAT 2 and 9 x S A T 3). All three SAT type viruses were frequently isolated on the same occasion from individual separate herds. In addition to titres against the SAT viruses, high SN titres up to 1 in 1024 were also found against types O, A and Asia 1 virus in sera from these buf­ falo (Table II) and results were very similar to those from Mozambique (Table III). The SN titres of individual animals frequently bore little relaTABLE II Reciprocal SN antibody titres and virus isolations, a selected group from 73 buffalo in the Kruger National Park. Animal No. SAT 1 SAT 2 SAT 3 F2 11 14 16 17 20 21 23 24 27 28 36 39 45 50 60 66 22 708 90 128 256 32 178 178 64 178 256 6 355 128 1024 90 178 355 355 64 90 355 355 256 256 1400 178 178 512 708 708 16 128 178 1024 1400 355 90 128 45 11 178 178 32 256 6 178 90 355 355 1400 Antibody titres O A C 64 256 22 8 16 6 6 22 8 178 64 45 16 16 16 22 128 178 6 1024 90 178 45 256 90 45 178 256 22 178 1024 22 178 512 6 11 6 6 45 6 6 64 6 6 6 32 6 32 6 6 45 Asia 1 Virus isolated 45 64 64 32 90 178 90 128 256 16 45 128 355 45 32 90 355 SAT 2 SAT 1 SAT 3 NVD* SAT 1 SAT 2 SAT 1 SAT 2 NVD SAT 1 SAT 3 SAT 2 SAT 1 SAT 2 SAT 3 NVD SAT 3 * N o virus detected. TABLE III Reciprocal log geometric mean SN antibody titres of buffalo in Mozambique and the Kruger National Park. 10 Survey Areas Mozambique* Kruger National Park** * 20 sera. ** 73 sera. SAT 1 Geometric mean antibody titre SAT 2 SAT 3 A 0 C Asia 1 2.46 1.78 2.56 1.52 1.71 1.21 1.86 1.98 2.30 2.12 1.31 1.76 0.99 1.80 — 391 — tionship to the virus type isolated and in many cases, heterologous titres, including those to the exotic types O, A and Asia 1 viruses, were as great or greater than the homologous (Table II). Among the exotic virus types SN titres were highest and most frequent against types A and Asia 1, less fre­ quent against type O and of a low order against type C virus. A collection of sera had been made from a free-living population of buf­ falo in north-western Botswana which had been sampled annually for FMDV over a nine-year period from 1968 to 1976. Four hundred and sixty virus iso­ lations had been made over this period and in every case the virus had been identified as either type SAT 1, 2 or 3. Representative groups of sera pre­ viously only tested for antibodies to SAT viruses were now tested against the exotic types O, A, C and Asia 1 viruses. Results were broadly similar to those from Mozambique and South Africa. In all groups some of the animals had SN titres to types O, A and Asia 1 viruses as high or of the same order as those against the SAT virus types. Similar titres were recorded also to type C in some animals. DISCUSSION Much of the field research into the role of African wildlife in the spread and maintenance of FMDV has been carried out in those parts of southern Africa where SAT virus types only have been recorded. Consequently, sera have usually been tested only against these virus types. The finding of high SN titres to both types O and A in buffalo in Mozambique was unexpected for although many hundreds of virus isolations have now been made from free-living buffalo in many African countries, there is no record of virus types other than SAT types being isolated. When the sera were tested against types C and Asia 1 virus, similar high titres were found to Asia 1. It was diffi­ cult to accept that infection with type Asia 1 virus, a type never previously recorded in Africa, had occurred in free-living buffalo. No samples for virus isolation had been taken. In a parallel study in South Africa's KNP O / P samples had been taken. Serological results from the sampled buffalo were similar to those from Mozambique but only type SAT viruses were isolated. It is known that following infection, FMDV persists in buffalo for very long periods (Hedger, 1976), but the sampled animals were young and infection would have had to have been recent. It would therefore be expected that, if infection had taken place with either types O, A or Asia 1 virus, then at least some of the virus isolations made would have been of these types. It may be argued that as sampling in the KNP was at one particular period of time, possibly only the prevalent virus types, well adapted to buf­ falo, had persisted, and that the antibody to types O, A and Asia 1 viruses may still have been evidence of previous but short-lived infection with these types. — 392 — Sampling in Botswana, however, had been taking place over a considera­ ble period of time (9 years). Up to 200 buffalo of a single population had been sampled annually and large numbers of virus isolations, all SAT types, were made. Sera from many of these animals had, in addition to antibody against the SAT virus types, high titres also to types O, A and Asia 1 viruses. A few had similar titres to type C virus. If these titres were reflective of past infection, it would be highly likely that, over this period of time, some isola­ tions of types O, A, C and Asia 1 viruses would have been made. When completely susceptible buffalo have been infected under experi­ mental conditions with a single FMDV type they have exhibited a monotypic antibody response (Young et al., 1972), but in almost all cases when freeliving buffalo populations have been sampled more than one SAT virus type has been isolated from the same herd. On a number of occasions individual buffalo have been found to be carrying more than one virus type and others, when sampled at a later date, have been carrying a type different from that originally isolated. By the age of three years, the majority of buffalo have probably been infected with all three SAT virus types. Although the reason for the high SN titres in buffalo to the exotic virus types O, A, Asia 1 and C must remain speculative, it is concluded they are not the result of infection with those particular viruses but are heterotypic responses following multiple infections with other prevalent virus types, i.e. the SAT viruses. Heterotypic responses occur in cattle following repeated infections with different virus types (Cottral and Gailiunas, 1971 : Hedger et al., 1980) and after vaccination following earlier infection (Hedger, 1967). It is possible they may also occur after infection of previously vaccinated animals. These observations emphasize the difficulties which attend the interpreta­ tion of serological results in FMD particularly in the absence of a known his­ tory of disease or vaccination. Retrospective diagnosis based on serology or serological surveys to determine virus type prevalence could be totally mislea­ ding in areas where FMD is enzootic or infection with more than one virus type has occurred. When fully susceptible animals, however, are infected for the first time, serological results are unequivocal and, in the absence of vaccination, the presence of significant amounts of antibody, albeit to multiple virus types, does at least indicate past infection and can be used in that context. It may also be possible, by examining a number of sera of unknown history against all the possible virus types, to exclude by comparison of titres, the likelihood of certain types having occurred. For example, examination of the serological data presented here would, on its own, suggest the exclusion of type C, where almost all animals have titres less than 1 in 64, from having occurred in the buffalo in southern Africa. — 393 — PRUEBAS SEROLÓGICAS RELATIVAS A LA FIEBRE AFTOSA REALIZADAS EN MUESTRAS DE SUERO BOVINO. PROBLEMAS DE INTERPRETACIÓN. — R.S. Hedger, I.T.R. Barnett, D.V. Gradwell y Paula Travassos Dias. Resumen : Cuando se infectan por primera vez con la fiebre aftosa animales totalmente receptibles, los resultados de las pruebas serológicas suelen carecer de ambiguedad. Surgen problemas cuando los animales fueron infectados o vacunados precedentemente con otros tipos de virus aftosos. La observación, en búfalos africanos salvajes de Mozambique y Africa del Sur, de títulos de seroneutralización frente a los tipos virales O, A y Asia 1 iguales o superiores a los títulos obtenidos frente a los tipos SA T, fue inesperada y condujo a otras encuestas. Aunque sea hipotética la causa de tan elevados títulos, cabe concluir que no eran el resultado de la infección por estos tipos especiales, sino que constituían respuestas heterólogas a raíz de infecciones múltiples producidas por otros tipos dominantes, sobre todo los virus SAT. Se discute la interpreta­ ción de los resultados serológicos por carecer de antecedentes conocidos de enfermedad o de vacunación. REFERENCES 1. CASEY Helen L. (1965). — Public Health Monograph, 74, 1-34. 2. CoTTRALL G.E. and GAILIUNAS P. (1971). — Experimental multiple infection of animals with FMD viruses. Proc. 75th Ann. Meeting, U.S. Animal Hlth Assoc. 1971. 3. GOLDING M., HEDGER R.S., TALBOT P. and WATSON J. (1976). — Radial immu­ nodiffusion and serum neutralisation techniques for the assay of antibodies to swine vesicular disease. Res. vet. Sci., 20, 142-147. 4. HEDGER R.S. (1967). — Field studies of FMD in Bechuanaland. Report of the Meeting of the Research Group of the Standing Technical Committee. European Commission for the Control of FMD. FAO, Rome. 5. HEDGER R.S. (1968). — The isolation and characterisation of foot-and-mouth disease virus from clinically normal herds of cattle in Botswana. J. Hyg. Camb., 66, 27-36. 6. HEDGER R.S. (1976). — Foot-and-mouth disease in wildlife with particular refe­ rence to the African buffalo (Syncerus caffer). In : « Wildlife Diseases » ed. L.A. Page, New York and London, Plenum Press. 7. KÄRBER G. (1931). — Beitrag zur kollectiven Behandlung pharmakologischer Reikenversuche. Arch. exp. path. Pharmak., 162, 480-483. 8. YOUNG E., HEDGER R.S. and HOWELL P.G. (1972). — Clinical foot-and-mouth disease in the African buffalo (Syncerus caffer). Onderstepoort J. vet. Res., 39 (3), 181-184.